📝 docs: logout does not invalidate server-side (live probe 2026-10-27) #162
Loading…
Reference in a new issue
No description provided.
Delete branch "docs/logout-not-server-side"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Summary
Today's live probes (while scoping #158) refuted a standing assumption:
POST /v1/logoutresponds 200 withSet-Cookie: cwt_session=; Expires=1970(client-side clearing only), and the old cookie value keeps working afterwards —GET /users/authenticated→ 200,GET /sites→ 200, and a fresh WS handshake is accepted. Thecwt_sessionJWE (alg: dir) appears stateless: there is no server-side session to revoke, the cookie lives until its embedded expiry (thex-cwt-tokeninstant). The docstring, README and field notes now say what was actually observed, including the security implication: a leaked cookie is valid until expiry — only time kills it.The client behavior itself is unchanged and remains sensible:
logout()drops the cookie from the jar and clears stored credentials soauto_reauthcannot silently restore the session.Evidence
pytest= 197 passed,mypyclean (docstring/README/field-notes only).Merge Danger
Door: two-way — documentation only.
Blast Radius: none in code; corrects a security-relevant claim in the docs.
WIP: 📝 docs: logout does not invalidate server-side (live probe 2026-10-27)to 📝 docs: logout does not invalidate server-side (live probe 2026-10-27)